| { | |
| "best_metric": 1.3191989660263062, | |
| "best_model_checkpoint": "./outputs/checkpoint-4000", | |
| "epoch": 2.9143897996357016, | |
| "eval_steps": 100, | |
| "global_step": 4000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3152, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 2.160670518875122, | |
| "eval_runtime": 143.4488, | |
| "eval_samples_per_second": 43.737, | |
| "eval_steps_per_second": 5.472, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1094, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 2.0639851093292236, | |
| "eval_runtime": 143.5664, | |
| "eval_samples_per_second": 43.701, | |
| "eval_steps_per_second": 5.468, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0347, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 2.0028491020202637, | |
| "eval_runtime": 143.5409, | |
| "eval_samples_per_second": 43.709, | |
| "eval_steps_per_second": 5.469, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9808, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 1.9539737701416016, | |
| "eval_runtime": 143.5859, | |
| "eval_samples_per_second": 43.695, | |
| "eval_steps_per_second": 5.467, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9262, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 1.911293625831604, | |
| "eval_runtime": 143.5587, | |
| "eval_samples_per_second": 43.703, | |
| "eval_steps_per_second": 5.468, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8974, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 1.8804633617401123, | |
| "eval_runtime": 143.5407, | |
| "eval_samples_per_second": 43.709, | |
| "eval_steps_per_second": 5.469, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8572, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 1.8469738960266113, | |
| "eval_runtime": 143.4729, | |
| "eval_samples_per_second": 43.73, | |
| "eval_steps_per_second": 5.471, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8337, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_loss": 1.8184312582015991, | |
| "eval_runtime": 143.5016, | |
| "eval_samples_per_second": 43.721, | |
| "eval_steps_per_second": 5.47, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7964, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_loss": 1.7926479578018188, | |
| "eval_runtime": 143.5332, | |
| "eval_samples_per_second": 43.711, | |
| "eval_steps_per_second": 5.469, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7803, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_loss": 1.7669757604599, | |
| "eval_runtime": 143.5614, | |
| "eval_samples_per_second": 43.703, | |
| "eval_steps_per_second": 5.468, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0002, | |
| "loss": 1.76, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 1.7429991960525513, | |
| "eval_runtime": 143.4985, | |
| "eval_samples_per_second": 43.722, | |
| "eval_steps_per_second": 5.47, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7218, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_loss": 1.7218060493469238, | |
| "eval_runtime": 143.5245, | |
| "eval_samples_per_second": 43.714, | |
| "eval_steps_per_second": 5.469, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7297, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 1.7012412548065186, | |
| "eval_runtime": 143.4698, | |
| "eval_samples_per_second": 43.73, | |
| "eval_steps_per_second": 5.472, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6814, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 1.679748773574829, | |
| "eval_runtime": 143.4782, | |
| "eval_samples_per_second": 43.728, | |
| "eval_steps_per_second": 5.471, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6509, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_loss": 1.6615930795669556, | |
| "eval_runtime": 143.519, | |
| "eval_samples_per_second": 43.715, | |
| "eval_steps_per_second": 5.47, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6331, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_loss": 1.6430609226226807, | |
| "eval_runtime": 150.5935, | |
| "eval_samples_per_second": 41.662, | |
| "eval_steps_per_second": 5.213, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6298, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_loss": 1.6253100633621216, | |
| "eval_runtime": 143.5003, | |
| "eval_samples_per_second": 43.721, | |
| "eval_steps_per_second": 5.47, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5992, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_loss": 1.6090008020401, | |
| "eval_runtime": 143.4499, | |
| "eval_samples_per_second": 43.737, | |
| "eval_steps_per_second": 5.472, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5908, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_loss": 1.5939785242080688, | |
| "eval_runtime": 143.55, | |
| "eval_samples_per_second": 43.706, | |
| "eval_steps_per_second": 5.468, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5723, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_loss": 1.5781009197235107, | |
| "eval_runtime": 143.4468, | |
| "eval_samples_per_second": 43.737, | |
| "eval_steps_per_second": 5.472, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5574, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "eval_loss": 1.5617916584014893, | |
| "eval_runtime": 146.1644, | |
| "eval_samples_per_second": 42.924, | |
| "eval_steps_per_second": 5.371, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5299, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 1.5464540719985962, | |
| "eval_runtime": 143.5449, | |
| "eval_samples_per_second": 43.708, | |
| "eval_steps_per_second": 5.469, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5286, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_loss": 1.5316085815429688, | |
| "eval_runtime": 143.5896, | |
| "eval_samples_per_second": 43.694, | |
| "eval_steps_per_second": 5.467, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5271, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_loss": 1.5180491209030151, | |
| "eval_runtime": 143.4857, | |
| "eval_samples_per_second": 43.726, | |
| "eval_steps_per_second": 5.471, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4973, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "eval_loss": 1.5037126541137695, | |
| "eval_runtime": 146.268, | |
| "eval_samples_per_second": 42.894, | |
| "eval_steps_per_second": 5.367, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.0002, | |
| "loss": 1.485, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "eval_loss": 1.490273118019104, | |
| "eval_runtime": 143.5853, | |
| "eval_samples_per_second": 43.695, | |
| "eval_steps_per_second": 5.467, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4731, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_loss": 1.4760063886642456, | |
| "eval_runtime": 143.5661, | |
| "eval_samples_per_second": 43.701, | |
| "eval_steps_per_second": 5.468, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4426, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_loss": 1.4635066986083984, | |
| "eval_runtime": 143.5367, | |
| "eval_samples_per_second": 43.71, | |
| "eval_steps_per_second": 5.469, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4152, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_loss": 1.4502085447311401, | |
| "eval_runtime": 143.5067, | |
| "eval_samples_per_second": 43.719, | |
| "eval_steps_per_second": 5.47, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4147, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "eval_loss": 1.4365726709365845, | |
| "eval_runtime": 143.5627, | |
| "eval_samples_per_second": 43.702, | |
| "eval_steps_per_second": 5.468, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3954, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "eval_loss": 1.4254179000854492, | |
| "eval_runtime": 143.4268, | |
| "eval_samples_per_second": 43.744, | |
| "eval_steps_per_second": 5.473, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4013, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "eval_loss": 1.414027214050293, | |
| "eval_runtime": 143.518, | |
| "eval_samples_per_second": 43.716, | |
| "eval_steps_per_second": 5.47, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3785, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_loss": 1.4006284475326538, | |
| "eval_runtime": 143.6131, | |
| "eval_samples_per_second": 43.687, | |
| "eval_steps_per_second": 5.466, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3608, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "eval_loss": 1.3891088962554932, | |
| "eval_runtime": 143.6908, | |
| "eval_samples_per_second": 43.663, | |
| "eval_steps_per_second": 5.463, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3617, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "eval_loss": 1.377625584602356, | |
| "eval_runtime": 143.5393, | |
| "eval_samples_per_second": 43.709, | |
| "eval_steps_per_second": 5.469, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3491, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "eval_loss": 1.3661019802093506, | |
| "eval_runtime": 143.5994, | |
| "eval_samples_per_second": 43.691, | |
| "eval_steps_per_second": 5.467, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3313, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "eval_loss": 1.3542754650115967, | |
| "eval_runtime": 143.6853, | |
| "eval_samples_per_second": 43.665, | |
| "eval_steps_per_second": 5.463, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3058, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "eval_loss": 1.343408226966858, | |
| "eval_runtime": 143.6047, | |
| "eval_samples_per_second": 43.689, | |
| "eval_steps_per_second": 5.466, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3154, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "eval_loss": 1.3290683031082153, | |
| "eval_runtime": 143.5227, | |
| "eval_samples_per_second": 43.714, | |
| "eval_steps_per_second": 5.47, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2978, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "eval_loss": 1.3191989660263062, | |
| "eval_runtime": 143.6032, | |
| "eval_samples_per_second": 43.69, | |
| "eval_steps_per_second": 5.466, | |
| "step": 4000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4116, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 2.3815364444912026e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |